AI Companies Are Picking Sides

Welcome back apprentices! 👋

Hey {{first_name|friend}},

For a while, AI felt like a group project where everyone was doing a bit of everything… and no one really knew who was in charge.

This week, that changed.

Some companies doubled down, others hit reset, and a few quietly made moves you probably didn’t notice (but should).

It’s starting to look more like a game where the rules just got real.

_{Read Time: 4 minutes}

🐶 Good Boy, Better Science. A dog with months to live just got a custom AI-designed cancer vaccine — built by her owner using ChatGPT, AlphaFold, and a genomics lab to turn raw tumor data into treatment. One tumor has already shrunk by 50%, proving this wasn’t just a science fair project but a real therapeutic swing. The wild part? No medical degree required — just tools that didn’t exist (or weren’t usable) a year ago.

🎸 The Band That Wasn’t (Until It Was). An AI-generated metal band racked up 80K listeners with fake bios, fake members, and very real fans—until Reddit noticed the hands looked… suspicious. Instead of disappearing, the creator hired real Tokyo musicians to bring the AI tracks to life on stage, turning a digital illusion into a live act. It’s weird, a little deceptive, and oddly effective — like Gorillaz meets ChatGPT.

🔓 AI’s “Oops, No Password” Moment. A security startup’s AI agent broke into McKinsey’s internal AI tool “Lilli” in under two hours — no Hollywood hacking, just exposed APIs with zero authentication. It unlocked tens of millions of sensitive messages, files, and user accounts (all in plain text), before the firm patched the issue and confirmed no wider breach. And this wasn’t a scrappy startup — it was McKinsey, where 45,000 employees rely on the tool daily.

The best AI outputs come from detailed prompts. But typing long, context-rich prompts is slow - so most people don't bother.

Wispr Flow turns your voice into clean, ready-to-paste text. Speak naturally into ChatGPT, Claude, Cursor, or any AI tool and get polished output without editing. Describe edge cases, explain context, walk through your thinking - all at the speed you talk.

Millions of people use Flow to give AI tools 10x more context in half the time. 89% of messages sent with zero edits.

Works system-wide on Mac, Windows, iPhone, and now Android (free and unlimited on Android during launch).

Try Wispr Flow free →

This week in AI felt like a strategic reset across the board. 

OpenAI is refocusing after realizing it may have spread itself too thin, Microsoft is restructuring to chase superintelligence (while fixing Copilot’s lukewarm adoption), Nvidia is doubling down on owning the entire AI infrastructure stack, and xAI is… basically starting over. 

Meanwhile, Google is quietly winning distribution by embedding Gemini everywhere, and MiniMax is pushing toward self-improving models that rewrite their own code. 

The common thread? Everyone’s choosing a core battlefield — coding, enterprise, infrastructure, or scale — and committing hard. 

The messy experimentation phase is ending, and the real competition is getting sharper, more focused, and a lot more serious.

OpenAI’s biggest threat right now is enterprise adoption. 

Anthropic’s Claude has quietly become a favorite for business workflows, especially coding, forcing OpenAI to confront a classic innovator’s dilemma: too many “cool” side projects (Sora, hardware, ads) vs. one core revenue driver. 

Simo calling it a “code red” internally signals a shift from exploration to execution. 

The Codex growth is promising, but the deeper issue is focus — enterprise buyers don’t want experiments, they want reliability and ROI.

Microsoft’s situation is more paradoxical: massive distribution (Office, Windows) but weak AI engagement. 

Copilot’s low adoption shows that bundling AI isn’t enough — it has to be genuinely useful. 

The reorg suggests Microsoft believes the real fix starts deeper, at the model layer, not just UX tweaks. And with fewer constraints from OpenAI, it’s now playing its own long game toward AGI. 

Translation: short-term product struggles, long-term ambition intact.

While others fight over models and apps, Nvidia is building the roads everyone drives on. 

Its strategy is deceptively simple: own chips, enable agents, power robotics, and stay open enough that everyone builds on you anyway. 

The “vertically integrated but horizontally open” line is key — it wants to dominate infrastructure without limiting adoption. If AI is the new electricity, Nvidia is positioning itself as the grid.

xAI is going through what looks like a controlled demolition. 

Losing 9 of 11 co-founders is definitely a reset. 

Musk admitting Grok is behind in coding is notable because that’s now the most competitive battleground. Hiring top talent from Cursor suggests a rebuild around developer tools, but the timing (pre-IPO) adds pressure. 

This is high risk: either a dramatic comeback or a very public struggle to keep up.

Google’s play is the least dramatic — and maybe the smartest. 

Instead of chasing headlines, it’s embedding Gemini into products people already use daily. 

Maps is a perfect example: adding AI without requiring behavior change. No new app, no learning curve — just better results. 

While others fight for users, Google starts with billions. That’s not a feature advantage; it’s a moat.

MiniMax’s M2.7 points to a bigger shift: models that improve themselves. 

Instead of humans iterating on training, the model writes code, evaluates itself, and loops. 

A 30% performance boost from this approach is early — but significant. 

If this trend holds, future AI development becomes less about training models and more about supervising systems that train themselves. That’s a very different game.

Today’s article is a map of where AI is heading. 

If you’re building, the signal is clear: pick a lane. Coding tools, enterprise workflows, infrastructure, or distribution — everyone winning right now is focused. 

If you’re a user, expect AI to feel less like a novelty and more like a utility (especially in tools you already use). 

And if you’re watching the space, the next phase won’t be about “who has AI,” but who makes it actually useful — and at scale.

So if your AI strategy still sounds like “we’re experimenting,” congratulations — you’re exactly where everyone was last year.

This guide distills 10 AI strategies from industry leaders that are transforming marketing.

Outpace your competitors by mastering AI.

Get Your Free Guide

Hey, you didn’t get all this info for nothing — share it! If you know someone who’s diving into AI, help them stay in the loop with this week’s updates.

Sharing is a win-win! Send this to a friend who’s all about tech, and you’ll win a little surprise 👀

Institutional giants use the $4 trillion rental market to compound millions. Now you can too. mogul offers fractional ownership in elite rental properties with 18.8% average IRR and zero property management required. Secure your share of the wealth Wall Street once kept for itself.

Skip The Line Now

Past performance isn't predictive; illustrative only. Investing risks principal; no securities offer. See important Disclaimers

A playground for your imagination (and low-key prompt skills).

Each send, we give you a customizable DALL·E prompt inspired by a real-world use case — something that could help you in your business or job if you wanted to use it that way. But it’s also just a fun creative experiment.

You tweak it, run it, and send us your favorite. We pick one winner to feature in the next issue.

Bonus: you’re secretly getting better at prompt design. 🤫

Last week, we challenged you to test GPT-4o’s visual generation skills with this prompt.

Here’s the WINNER:

Congrats to Jusef for his creation !🥳

Want to be featured next? Keep those generations coming!

At a busy city intersection, the pedestrian crosswalk has been completely transformed into a bold, high-impact visual. Instead of standard white stripes, the entire crossing is redesigned around [your pattern], stretching across the asphalt in vivid color and precise geometry. Cars wait at the lights while people walk across it, interacting naturally with the design. The paint is fresh and slightly reflective, catching sunlight and city glow, with crisp edges and ultra-detailed textures against the rough road surface. Shot from an elevated cinematic angle, with sharp clarity, strong contrast, and vibrant color — turning an everyday urban moment into a striking piece of visual identity.

If AI outputs feel inconsistent, it’s usually not the model, it’s missing structure.
We documented the exact 6- Part System we use to get reliable results across ChatGPT, Claude, and Gemini.

It’s a short guide you can finish in under an hour, with plug-and-play prompts + exercises so you actually build the skill and fix the frustrating AI inconsistencies.

Subscriber Price: $10 (normally $19).

_FEEDBACK